Reducing carbon emissions in the cloud

January 16, 2014 | By Barbara Vergetis Lundin

New research by computer scientists at Trinity College Dublin and IBM Dublin could allow companies to reduce greenhouse gas emissions and drive down costs of operations using mathematical algorithms to effectively balance the load between different computer servers worldwide. All of the services on the Internet today are based in the cloud, which means Twitter, Facebook or Google mail requests are dealt with by one of thousands of PC servers located at a small number of warehouse-sized cloud-computing facilities around the world.

The cloud-computing facilities consume megawatts of power and generate a level of greenhouse gas emissions that varies depending on factors such as local time, the utility's fuel mix for electricity generation, and the use of sophisticated power-saving techniques. For example, a facility fed by a coal-fired generator on a hot summer day consumes power that is expensive and generates a high carbon output. In contrast, a facility operating next to a large wind farm in the middle of the night will be cheaper and more carbon efficient.

Companies that host their services in the cloud need to buy sufficient capacity to meet demand, but can choose where they want their servers to be located and can change this on an hourly basis. Known as Stratus, the system allows a company to set out how much importance they attach to cost, greenhouse gas emissions and network delays involved in servicing their internet load. The algorithms then work out how best to split the load across different cloud-computing facilities to achieve the best result.

"The overall goal of the Stratus system is to allow companies to procure their cloud-computing service in a way that best serves their priorities," said Donal O'Mahoney, professor of computer science at Trinity. "If they want to be super-green, it will shift the load one way. If they want to cut costs to the bone, it will shift it another way, or they can choose anything in between."

To test the system, the scientists created a simulation based on three large facilities in California, Virginia and Dublin, which mirrored where Amazon Web Services -- a leading cloud provider -- has three of their major service centers. Some data centers used state-of-the-art cooling techniques and some did not. Real-time local wholesale prices for electricity were used where available. The team also conducted a study on network delays between various parts of the Internet and the three facilities.

In their simulations, the scientists found that by tailoring the algorithms to reduce carbon output, they could achieve a 21 percent reduction in the greenhouse gas emissions associated with the given load. By targeting electricity cost reductions, they could achieve a 61 percent savings over simply splitting the load evenly. By assigning weights to each factor, the load could be spread to fully reflect individual preferences.

For more:
- see this report

http://www.fierceenergy.com/story/reducing-carbon-emissions-cloud/2014-01-16